Run 4. Outer Step 4. Inner Step 4.

config.json

@@ -268,18 +268,20 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5486468,
-    5486472,
-    5486475,
-    5486477,
-    5486480
+    5486484,
+    5486487,
+    5486488,
+    5486491,
+    5486494,
+    5486496,
+    5486499
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 3,
+  "inner_step": 4,
   "inner_steps": 0,
   "last_allreduce_block": 5485540,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4a5c543a8c0d5da23d62770777c0a7fb8466bf1c49fc23d85cc9c7235a225218
+oid sha256:5db729f49ab23050a5181a620110103ce90d95ff51950290a99809d3801f8d31
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7d2c6b0eda6556b503e7807071c99ed2f5584b0a338a09410f9b126ab0c135bd
+oid sha256:3921cdc12a54dfaba30734f90cb1d3540bf169aa1b75cf16783d8d38eda3d2d1
 size 4040701744